Solar desalination coupled with water remediation and molecular hydrogen production: A novel solar water-energy nexus

Transcription

1 Electronic Supplementary Material (ESI) for Energy & Environmental Science. This journal is The Royal Society of Chemistry 17 Supporting Information Solar desalination coupled with water remediation and molecular hydrogen production: A novel solar water-energy nexus Seonghun Kim, 1 Guangxia Piao, 1 Dong Suk Han, Ho Kyung Shon, 3 and Hyunwoong Park 1, * 1 School of Energy Engineering, Kyungpook National University, Daegu 41566, Korea Chemical Engineering Program, Texas A&M University at Qatar, Education City, P.O. Box 3874, Doha, Qatar 3 School of Civil and Environmental Engineering, University of Technology, Sydney, Post Box 19, Broadway, NSW, 7, Australia *To whom correspondence should be addressed (H. Park): School of Energy Engineering, Kyungpook National University, Daegu 41566, Korea hwp@knu.ac.kr; Tel: Figure S1. SEM images of TiO nanorod () arrays grown on FTO.

2 Intensity (a.u) Rutile, ICDD # (11) * (111) * 6h 5h 4h () * (31) (11) * * FTO h FTO Degree) Figure S. XRD spectra of samples hydrothermally grown for varying times (3 6 h) on FTO.

3 Intensity (a.u) a TiO b background c Ti p H- 3/ SnO TiO Ti-OH SnO Ti-OH Ti p 1/ TiO SnO Ti-OH H background TiO SnO Ti-OH Binding Energy (ev) Binding Energy (ev) Binding Energy (ev) Figure S3. XPS spectra of and H- samples: (a) Tip and (b and c) O1s. Upon harsh reductive conditions, white TiO turns blue or black depending on the degree of the fraction of Ti(III) or oxygen vacancies. Our samples with a mild H treatment were gray, suggesting that only a small fraction of Ti(VI) was reduced and hence the Tip band shift was not found. Similarly to our observation, there are many reports that the mild H treatment did not shift the XPS Tip band (Nano Lett. 11 (11) 36; ibid. 14 (14) 339; Chem. Phys. Lett. 9 (198) 67). Instead, they revealed that the intensity of the surface oxygen coordinated with hydrogen (e.g., >Ti-OH) increased with increasing the oxygen vacancies. In typical, TiO exhibits two XPS O1s bands at ~59.5 ev and ~53 ev. The former is attributed to the lattice oxygen inside TiO bulk and not affected by the surface treatment of TiO, whereas the latter is associated with the surface oxygen (i.e., >Ti-OH) and significantly influenced by the surface environment. Based on this knowledge, we have deconvoluted the XPS O1s bands of and H- samples. As shown in Fig. S3, the O1s peak at ~53 ev in the H- was significantly stronger than that in the. This result indicates that the H- had a large degree of oxygen vacancies as a result of the reduction of Ti(IV) although the shift of the Tip peak was not obvious.

4 4 H- 3 I ph (ma/cm ) Time(h) Figure S4. Time-profiled photocurrents generated with and H- held at +.5 V vs. SCE under AM light (1 mw cm ) in aqueous sodium sulfate (.1 M). Figure S5. Time-resolved photoluminescence emission spectra of and H- in the wavelength range between 4 and 45 nm. The estimated lifetimes were summarized.

5 Figure S6. (a) Mott-Schottky plots and (b) Nyquist plots of and H- electrodes in an aqueous sodium sulfate (.1 M). In b, interfacial and internal charge transfer resistance values (R and R3) were estimated based on the electrochemical circuit model presented.

6 Comparison of and H- for their photoelectrocatalytic activities 6 1 a 8 1. b O ( mol) 4 FE (%) C t / C / RNO H-. H- / RNO / Formate H- / Formate Time (min) Figure S7. (a) Evolutions of O (inset: the Faradaic efficiencies for the evolved O ). (b) Decompositions of RNO and formate using and H- at +.5 V vs. SCE in aqueous sodium sulfate solutions (.1 M) under AM light (1 mw cm ). The photoelectrocatalytic production of molecular oxygen (O ) proceeds via water oxidation (R1). H O + 4h + O + 4H + E = 1.3 V (R1) As shown in Figure S7a, O was linearly produced over 3 h at the production rates of 14.9 and 17.8 mol h 1 for the and H-, respectively. Despite the higher O production rate and the larger photocurrent (Figure S4), the Faradaic efficiency of the H- was lower than that of the (~63% vs. ~8%). We speculated that the lower Faradaic efficiency of the H- for OER is attributed to other oxidation pathways including single-electron oxidations. To examine this possibility, the PEC oxidation of RNO (hydroxyl radical quencher) by hydroxyl radicals (R) was carried out using the and H- (Figure S7b). H O + h + OH + H + E =.73 V (R) The RNO oxidation was ~1.3-fold faster in the H- with the pseudo-first order reaction (C t /C = exp( k app t), where C t and C refer to the concentrations of RNO at times zero and t, respectively) rate constant (k app,h- ) of 1 min 1 (c.f., k app, = min 1 ). Although this suggests that the hydrogen treatment of enhances the production of hydroxyl radicals, the difference is not significant. The similar kinetics in the PEC oxidation of phenol (Figure S8) via the mediation of hydroxyl radical confirmed the insignificant effect of the hydrogen treatment on the R pathway. We further speculated the existence of direct electron transfer and employed formate as an alternative substrate to examine the pathway

7 (R3). HCOOH + h + COOH + H + + e (R3) As shown in Figure S7b, the formate concentration decreased linearly with time and the degradation rate with H- was ~1.6-fold larger than that with (k app,h- = min 1 ; k app, = min 1 ). The enhancement factor for the formate decomposition by the hydrogen treatment is greater than that for RNO. This suggests that H- can be utilized for the various single-electron transfer oxidation processes including reactive chlorine species (RCS)-mediated oxidation (see R4 R7). Cl + h + Cl E =.41 V (R4) Cl + Cl Cl (R5) Cl + H O + h + HOCl + H + E = 1.49 V (R6) RCS + Urea Decomposition + xcl (R7) C t /C (PhOH) HQ ( M) 1,3,5-HB ( M) C =.1 mm Hydroquinone (1OH) 1,3,5-Hydroxylbenzene (OH) Time (min) H CC ( M) RC ( M) Conc ( M) Catechol (1OH) Resorcinol (1OH). 1 8 PhOH 6 4 New OH in intermediates Time (min) Figure S8. Photoelectrocatalytic decomposition of phenol (PhOH,.1 mm) and concomitant productions of hydroxylated intermediates (hydroquinone, catechol, resorcinol, and 1,3,5- hydroxylbenzene) using and H- photoanodes held at +.5 V vs. SCE in aqueous sodium sulfate solutions (.1 M) under AM light (1 mw cm ).

8 8 Free chlorine ( M) 6 4 H Time (min) Figure S9. Photoelectrocatalytic productions of free chlorine using and H- photoanodes held at +.5 V vs. SCE in aqueous sodium chloride solutions ( mm) under AM light (1 mw cm ). Figure S1. Photoelectrocatalytic urea (C = 1 mm) decomposition using H- at +.5 V vs. SCE in the anode cell under AM light (1 mw cm ) and simultaneous H productions in the cathode cell. Both cells were filled with the same solution of mm chloride and separated by a proton-exchange membrane.

9 Figure S11. Time-profiled ion transport efficiency (= Amount of monovalent ion (mol) for t / (I ph t)) using and H- photoanodes in three-cell stack. See Figure 4 for timeprofiled changes in the amounts of ions and I ph ph Anolyte, -compartment cell Catholyte, -compartment cell Anolyte, 3-compartment cell Catholyte, 3-compartment cell Figure S1. Time-profiled ph changes in anolyte and catholyte in the two-compartment cell with PEM and three-compartment cell with AEM and CEM. In the three-compartment cell, a decrease in anolyte ph is attributed to the oxidations of water and chloride (R1 and R6, respectively), whereas ph increase in catholyte results from H production via proton reduction (H + + e H O). This ph difference was not found in the twocompartment cell because of proton transport.

10 1. Total chlorine Free chlorine a Chlorine (mm) b Nitrogen (mm) 3 1 Urea () Intermediates (1) Nitrate (1) Ammonia (1) Deficient N (1) Figure S13. (a) Time-profiled concentrations of total and free chlorines. Total chlorines refer to reactive chlorine species bound to and free from urea. (b) Time-profiled nitrogen balance. The numbers in parentheses indicate the number of nitrogen atom in the corresponding compounds. Intermediates refer to mostly chloroamines (NH x Cl 3-x, where x =, 1,, 3) which were estimated based on the difference of total and free chlorines. Most of the deficient nitrogen should be attributed to the production of N.

11 Figure S14. Time-profiled potential changes of photoanode (E a for held at +.5 V vs. SCE) and cathode (E c for Pt foil) in the three-cell stack. E stack refers to the overall operational stack voltage. See Figure 3 for the corresponding reaction. -1 Before reaction After reaction ~ V vs. RHE I (ma/cm ) E (V vs. SCE) Figure S15. Comparison of linear sweep voltammograms of Pt electrodes before and after desalination (for 1 h). Electrolyte for LSVs: aqueous NaOH solution (.1 M, ~ph 13).

12 I ph (ma) or E stack (V) Urea (mm) Products (mm) E for H- E for I ph for H- a H- I ph for / Nitrate / Ammonia H- / Nitrate H- / Ammonia c Ions (mmol) H ( mol) Anolyte & Catholyte Middle cell b (Na) (Cl) H- (Na) H- (Cl) (Na) (Cl) H- (Na) Figure S16. Photoelectrocatalytic urea decomposition by and H- in the anode cell containing urea (C = 1 mm, 15 ml), desalination in the middle cell containing NaCl (C = 17 mm, 1 ml), and simultaneous H productions in the cathode cell containing mm K SO 4 (15 ml). I ph refers to photocurrents produced from and H- held at +.5 V vs. SCE under AM light (1 mw cm ) and E stack indicates the overall stack voltage between the photoanode and the cathode (Pt). (a) Changes in I ph and E stack. (b) Changes in the amounts of ions in the anode, cathode, and middle cells. (c) Time-profiled changes in urea concentration (upper panel) and concomitant productions of ammonia and nitrate (lower panel). (d) H productions in the cathode cell. 3 1 H- H- (Cl) d

13 I ph (ma) H a Ions Anolyte & Catholyte (mmol) Na in CA Cl in An Na in CA Cl in AN Na in MD Cl in MD. 1. Na in MD.5 Cl in MD Void symbols: Filled symbols: H- b Ions middle (mmol) Products (mm) Urea (mm) Nitrate Ammonia Nitrate Ammonia H- Circles: Traiangles: H c H ( mol) H- (FE) H- (FE) d Faradaic Efficiency (%) Figure S17. Photoelectrocatalytic urea (C = mm, 15 ml) decomposition using and H- in the anode cell, desalination in the middle cell containing NaCl (C = 17 mm, 1 ml), and simultaneous H productions in the cathode cell containing K SO 4 ( mm, 15 ml). I ph refers to photocurrents produced from H- (or ) photoanode and Pt cathode couples held at E stack of V under AM light (1 mw cm ). (a) Changes in I ph and E stack. (b) Changes in the amounts of ions in the anode, cathode, and middle cells. (c) Time-profiled changes in urea concentration (upper panel) and concomitant productions of ammonia and nitrate (lower panel). (d) H productions in the cathode cell and Faradaic efficiency for the produced H.